perl
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues detected. The skill provides educational content, coding examples, and best practices for Perl development including features from v5.36 to v5.40.\n- [INDIRECT_PROMPT_INJECTION]: The skill documents techniques for parsing external, unstructured data such as log files and data streams, representing a standard indirect prompt injection surface.\n
- Ingestion points: File parsing templates in
SKILL.mddemonstrate processing external lines via filehandles or string variables.\n - Boundary markers: The provided regex examples provide data extraction without specific delimiters or instructions for the agent to ignore potentially embedded commands in the data.\n
- Capability inventory: The documented capabilities are limited to standard Perl text manipulation and console output (
say).\n - Sanitization: The provided examples demonstrate extraction but do not explicitly show sanitization or escaping of the content for downstream consumption.
Audit Metadata